City Research Online

Abstract

This thesis examines the dry bulk sector of the shipping industry. We begin by analysing the relation between second-hand vessel prices, net earnings, and holding period returns. Specifically, we provide strong statistical evidence that almost the entire volatility of shipping earnings yields can be attributed to variation in expected net earnings growth; almost none to expected returns variation and almost none to varying expectations about the terminal earnings yield. According to our results, earnings yields are negatively and significantly related to future net earnings growth. Furthermore, we find no consistent, strong statistical evidence supporting the existence of time-varying risk premia in the valuation of dry bulk vessels. Accordingly, we integrate the examination of the second-hand market by incorporating in the analysis the trading activity related to dry bulk vessels. For this purpose, we develop a heterogeneous expectations asset pricing model that can account for the actual behaviour of vessel prices and the positive correlation between net earnings, vessel prices, and second-hand vessel transactions. The proposed economy consists of two agent types who form heterogeneous expectations about future net earnings and at the same time under(over)estimate the future demand responses of their competitors. Formal estimation of the model suggests that the average investor expectations in the second-hand market for ships must be “near-rational”. In particular, the investor population must consist of a very large fraction of agents with totally – or very close to – rational beliefs while the remaining ones must hold highly extrapolative beliefs; thus, there must exist significant heterogeneity of beliefs in the market. Having concluded the analysis of the second-hand physical shipping market we turn to the derivative market for Forward Freight Agreements (FFAs) related to the dry bulk shipping sector. Accordingly, we illustrate formally that the bulk of volatility in the FFA basis can be attributed to expectations about future physical market conditions rather than expectations about future risk premia. Despite this finding, though, we document the existence of a bias in the FFA rates in the form of “contango” but also of both a strong momentum effect and significant predictability of risk premia by price-based signals and economic variables reflecting physical market conditions. The evidence of bias is further supported by the results of three econometric tests which suggest rejection of the unbiased expectations hypothesis. Finally, to justify these findings, we develop a dynamic asset pricing framework that can incorporate both the “hedging pressure” feature and a heterogeneous-beliefs explanation.

Publication Type:	Thesis (Doctoral)
Subjects:	H Social Sciences > HG Finance
Departments:	Bayes Business School Doctoral Theses Bayes Business School > Bayes Business School Doctoral Theses

Export

Downloads